Journal of Irrigation and Drainage Engineering

The height, number of tillers, chlorophyll (Chl) a and b content, and yield of rice (Oryza sativa L.) in response to alternating stresses of drought and waterlogging (ASDW) at the tillering and jointing stages were evaluated in a greenhouse. This was done to simulate field conditions and explore appropriate water management strategies, including the use of rainwater. The experiment tested four different ASDW treatments and regular irrigation (CK) in 2012 and 2013. The ASDW treatments were light drought + light waterlogging + light drought (T-LD) and severe drought + light waterlogging + severe drought (T-HD) at the tillering stage, and light drought + light waterlogging + light drought (S-LD) and severe drought + light waterlogging + severe drought (S-HD) at the jointing stage. CK was the control. Plant heights at harvest stage with the T-HD and T-LD treatments were 7.35 and 8.65 cm higher than plant heights with CK in 2012, while plant heights with the S-HD and S-LD treatments were lower than plant heights with CK in 2012 and 2013. Compared with CK, the number of tillers with the T-LD and T-HD treatments was reduced by approximately 1 to 2, and the number of ineffective tillers in the S-LD and S-HD treatments was reduced by approximately 2 to 3 and 2 to 4, respectively. After ASDW with the T-LD and T-HD treatments the Chl a and b content increased in 2012, but with the S-LD and S-HD treatments the Chl a and b content decreased by 8.11% and 10.81%, respectively, in 2013. Decreases in the 1,000-grain weight and filled grains per spike were the main causes of yield reduction. Yield per pot with the T-LD and T-HD treatments decreased by 29.25% and 25.37%, respectively, in 2013 ($P<0.05$). Yield per pot with the S-LD and S-HD treatments decreased by 19.01% and 39.80%, respectively, in 2013 ($P<0.05$). The harvest index values for the T-LD and T-HD treatments were higher than those for CK. The irrigation water use efficiency (IWUE) value for the T-HD treatments improved by 46.77% ($P<0.05$) in 2012, and the IWUE value for the S-HD treatments decreased by 29.78% in 2013.

Numerical one-dimensional (1D) model simulations were performed on the infiltration of freshwater as saturated plug flow into an unsaturated sandy soil at field capacity composed of coarse sand or sands (with saturated hydraulic conductivity, $K_s$, ranging from 0.004 to $0.1\mathrm{cm}/\mathrm{s}$ according to the USDA). Simulation used the Richards equation for the flow of water and an advection-dispersion equation for the transport of salt in aqueous solution. The 1D model had been validated previously against experimental laboratory data. Existing simulation models of transport in porous media, e.g., HYDRUS, are more powerful than what was needed in this study, therefore we chose to use our own model. In the regions where road salt is used, the saline pore water initially near the ground surface is from road salt (NaCl) applications during winter; in the following warm season, the soil thaws, the saline water percolates, and some amount of saline water is left in the pores of unsaturated soil. Infiltration of freshwater from rainfall can rinse accumulated saline water stored in the pores of the soil, the depth of which is up to 5 m. In time, the saline water will reach shallow groundwater and pollute aquifers because ${\mathrm{Cl}}^{-}$ can be toxic. The simulation results show that only a few hours of continuous infiltration are needed for highly permeable soil ($K_s$ from 0.01 to $0.1\mathrm{cm}/\mathrm{s}$) in order to reduce the salinity in the soil to less than 1% of its initial mass averaged across the entire modeling domain. For $K_s=0.004\mathrm{cm}/\mathrm{s}$, a few days (depending on the soil depth) of continuous freshwater infiltration is needed. The long required times indicate that the saline pore water is difficult to remove in one rainfall event from soils consisting of finer sands. The results are consistent with findings of significant salt concentrations in surface and shallow groundwater, long after road salt has been applied.

The common belief is that there are opportunities in the agricultural sector to use less irrigation water and thus make saved water available for municipal and industrial uses. The key is for farmers to maximize farm profit through optimizing water allocation between on-farm production and off-farm leases. A model to optimize water allocation for a single farm and crop using crop water production functions and the concept of deficit irrigation is presented in this paper. Using this model, optimal water allocation was determined based on maize and sunflower water productivity data collected during 2008–2011 growing seasons in eastern Colorado. According to the region’s water leasing and yield prices during 2010–2016, deficit irrigation does improve farm income at moderate to high leasing prices. When crop prices are low relative to leasing prices, leasing all the available water and fallowing or planting unirrigated crops provides the highest income. The model can define the crop and water leasing prices for which deficit irrigation is worthwhile.

The curve number (CN) method has been widely used to estimate the amount of direct runoff generated in agricultural watersheds around the world. However, its applicability to rice paddy fields has yet to be sufficiently tested to suggest the range of alternative CNs for irrigated and drained paddies. This study proposes CNs applicable to rice paddies that have different water management practices based on nine-year rainfall, irrigation, and drainage monitoring data. The results suggest that CNs should be adjusted to consider management practices characterized by rain-harvesting rates and drainage equipment, especially with a shallow ponding depth. When the soil water content was close to saturation and ponding depth was high due to irrigation and antecedent rainfall, CNs were less sensitive to drainage practices than they were under dry conditions. This study clearly demonstrated that irrigation and drainage practices implemented to manage rice paddies significantly affect the hydrological responses of the plots and subsequently impact CNs, suggesting the importance of careful CN determination for rice paddies. The proposed CNs are expected to be useful references in modeling hydrological processes in paddy rice areas and in developing agricultural watershed management plans using the CN method.